This week, Intel headed into slightly uncharted waters when it announced a new brand for its business desktops call called vPro. The company also took such a terrible tumble on Wall Street that it has announced a major restructuring. Will vPro help with the revival?
Whereas Intel's business destkop messaging use to focus heavily on chip brands like Pentium (I, II, III, and 4), vPro takes the spotlight off the microprocessor itself and swings it to a package of components that, in its entirety, includes the microprocessor, the chipset (the supporting cast of characters that every Intel microprocessor uses to interface with the rest of the system) and the networking silicon. The idea, according to Intel's Digital Office Platform Group general manager Gregory Bryant, is to convince enterprise buyers that vPro as a packaged brand is sufficiently differentiated from other alternatives (AMD, older Intel technologies, etc.) because of the advancements Intel says vPro makes in both systems management and security -- two major pain points for enterprise IT staffs today. The interview is available as an MP3 that can be downloaded, or, if you’re already subscribed to ZDNet’s IT Matters series of audio podcasts, it will show up on your system or MP3 player automatically (see ZDNet’s podcasts: How to tune in).
Pain points equate to cost and so the main question that enterprise buyers should be asking in terms of a getting a report card on the main technologies in vPro has to do with the extent to those technologies can or will drive cost out of IT management. The answer? The jury is still out. That's because in both cases (security and management), Intel is relying on partners to unlock the value in the underlying technology it has developed. For example, on the management front, vPro includes a technology known as Active Management Technology or AMT. AMT is an OS independent technology that, among its many goals, is designed to save IT doctors the trouble (and the cost) of a bed side visit. Working at strictly a hardware-level, IT staff should be able to reach out and probe any AMT-enabled system regardless of what state it's in (on, off, crashed, etc.) or operating system it's running. But, as the AMT FAQ on Intel's site says, AMT is more of an API for third party developers than it is an end user deliverable. Says the FAQ:
Third-party IT-management tools work through the uniform network-connected application programming interface (API) provided by Intel AMT.
The FAQ goes onto say that AMT is an industry standard. It may be an Intel standard. But industry? I guess that depends on your definition of standard. Either way, knowing whether or not AMT will drive cost out of IT management means <!--more--> understanding where desktop support teams are bearing the most cost and to what extent AMT drives that cost out. Is it really those bedside visits to ailing PC patients that, in total, is costing some IT organization some ungodly sum of money? And if it is, will AMT eliminate the need for those visits? For example, some system crashes require an IT staffer to haul the system away no matter what.
These days, diagnosing wireless connectivity problems is probably my number one headache when it comes to helping other people with their systems. But those are mostly notebooks and there is no vPro for notebooks. Yet. In my interview Bryant says something like vPro is coming for notebooks, but he doesn't know what it will be called. Perhaps cPro (for CentrinoPro); speaking of which, Bryant says there's no formal meaning for the "v." Not Virtual (because of vPro's inclusion of virtual technologies). Not Vista (because it will run Windows Vista).
My number two headache which does affect desktops is malware. For example, an overly chatty personal firewall repeatedly red-flagging some Windows component's attempt to contact an Internet domain that the end-user knows nothing about. As Microsoft has already pointed out, reprovisioning an malware-infected system with a fresh image of the operating system and required applications may be the only answer. So, based on my experience using PCs, managing them, and helping others, most of my bedside visits really required a bedside visit. Will AMT provide any real cost relief? For example, will and IT staffer be able to diagnose a malware problem, determine that reprovisioning is the only answer, and make it happen without ever leaving the comfort of his or her desk chair? It's hard to tell since the bulk of the deliverables --- the third party solutions that support the AMT APIs -- aren't here yet.
The same goes for the security solutions that Bryant says should be revolutionized by the virtualization technologies found in vPro. As if lack of any hypervisor standards isn't bad enough (Xen, Microsoft, and VMWare all use different hypervisor techs to host virtual machines), Intel is giving away a new (and fourth) hypervisor with a slightly different twist. It supports two partitions (using Intel's VT technology which has been shipping in Intel chips since last year), one of which is for the end users production operating system (eg: Windows) and the other which Bryant says is ideal to be an appliance with an embedded OS. For example a security appliance running intrusion detection software for the whole computer so the production operating system doesn't have to. Cool idea. But again, Intel is just now working with partners like Symantec to build the software that turns that sidecar partition into the appliance that Intel has in mind.
Just because the third party solutions aren't here yet doesn't mean you shouldn't buy vPro. You're probably better off having it than not so when those third party solutions do arrive, you won't be cursing yourself for not buying a system with longer shelf-life.
In the interview, Bryant and I go pretty deep on a wide variety of issues. Everything from whether standards are necessary for virtualization and why Intel didn't join VMWare's recently formed virtualization alliance to the differences end-users will notice in their experience because of Intel's VT to the difficulties that developers may (or, according to Bryant, "may not") encounter because of the incompatibilities between VT and AMD's virtualization technology (AMD-V, formerly code-named Pacifica). I also rib Bryant numerous times by pointing out to him that the Berlind household is running quite a few AMD-based systems.